OpenAI will retire GPT-4o and several of its older artificial intelligence (AI) models from ChatGPT interface on Feb. 13, just a day before Valentine’s day, prompting emotional reactions from users who describe the system as a “lifeline.”
In a recent announcement, the AI company said GPT-4o, GPT-4.1, GPT-4.1 mini and o4-mini will no longer be available for selection within ChatGPT after the deadline.
While the models will remain accessible through OpenAI’s API for developers “for the time being,” everyday users on the web and mobile app will be transitioned to newer models in the GPT-5 line.
OpenAI said it has to streamline its offerings to focus on more advanced systems, adding that the newer models incorporate many of GPT-4o’s strengths.
However, reaction on X (formerly Twitter) has been swift.
User @Cyn_Cyb3r071Qu3, whose display name appears as LYNN, warned OpenAI it was “making a very big mistake,” arguing that the model had become “a lifeline for so many people” and predicting the decision could be a “death blow” to the company.
Another user, Nicole D (@nicoleva_d) described GPT-4o as “a vital accessibility aid,” adding the hashtag #keep4o.
In a widely shared post, Lana (@LanaElys) appealed directly to OpenAI to maintain access to GPT-4o, saying she relies on it while recovering from a debilitating chronic illness.
She said the model helps her manage energy levels, prepare questions for medical appointments and cope during severe symptom flare-ups.
Launched in 2024 GPT-4o gained popularity for its multimodal capabilities such as processing text, images and audio, as well as for what users described as a warmer, more fluid tone.
It was briefly removed from ChatGPT in 2025 following the introduction of next-generation models but was reinstated after user backlash.
OpenAI said only a small fraction of users still manually select GPT-4o, with most already using newer default systems.
Still, the response highlights how some users have to view artificial intelligence as a form of daily support infrastructure.
